release(2.0.0): base-template retrain on posix-sdc v2.0.0

test(3): verify-then-exit corpus is a negative result; v1.1.0 stays champion

test(2): r=32+MLP capacity arm beats r=16 across the board

chore(2): add r=32 + MLP-targets retrain wrapper (capacity arm)

test(1): greedy (temp-0) re-eval; decoding exonerated, capacity is the lever

feat: commit trained adapter, QLoRA config, and held-out eval results